PROJECT TITLE :

Automatic Selection of Partitioning Variables for Small Multiple Displays

ABSTRACT:

Effective small multiple displays are created by partitioning a visualization on variables that reveal fascinating conditional structure in the information. We tend to propose a method that automatically ranks partitioning variables, permitting analysts to target the foremost promising small multiple displays. Our approach is based on a randomized, non-parametric permutation take a look at, which permits us to handle a wide selection of quality measures for visual patterns defined on many totally different visualization sorts, whereas discounting spurious patterns. We tend to demonstrate the effectiveness of our approach on scatterplots of real-world, multidimensional datasets.